如何在Rmarkdown中格式化multcomp包glht结果为HTML表格

summary.glht对象无法直接被pander/kable/flextable识别，核心解决思路是提取对象内的检验结果，转换为标准数据框后再处理。

可重复示例与提取步骤 #

# 加载包并构建示例模型
library(multcomp)
set.seed(123)

# 模拟数据
Habitat <- factor(rep(c("A", "B", "C"), each = 20))
Response <- c(rnorm(20, 5), rnorm(20, 7), rnorm(20, 9))
tick_oneway <- aov(Response ~ Habitat)

# 执行Tukey事后检验并生成summary
post_hoc <- glht(tick_oneway, linfct = mcp(Habitat = "Tukey"))
post_hoc_sum <- summary(post_hoc)

从post_hoc_sum中提取关键检验指标，组合为数据框：

# 提取对比项、系数、标准误、t值、p值
result_df <- data.frame(
  Contrast = rownames(post_hoc_sum$test$coefficients),
  Estimate = post_hoc_sum$test$coefficients,
  Std.Error = post_hoc_sum$test$sigma * sqrt(diag(post_hoc_sum$test$vcov)),
  t_value = post_hoc_sum$test$tstat,
  Pr_t = post_hoc_sum$test$pvalues
)

用表格工具生成HTML表格 #

1. knitr::kable

library(knitr)
kable(result_df, format = "html", caption = "Tukey Post Hoc Test Results")

2. flextable

library(flextable)
as_flextable(result_df) %>% 
  set_caption("Tukey Post Hoc Test Results") %>% 
  autofit()

3. pander

library(pander)
pander(result_df, caption = "Tukey Post Hoc Test Results")

内容的提问来源于stack exchange，提问作者Geosphere